Practical guide · September 2026 · about 6 minutes · the starting point for this series
AI gets sold to tradespeople as a revolution: agents, instant quotes, a business that runs itself. The useful version is a lot less dramatic. It’s help with the jobs around the job — the ones that decide whether you win the work and get paid for it.
The short version
- Most small businesses are still trying AI, not relying on it. Of US small employers using it, only 7% have it fully built in.
- Below are ten practical uses — from getting back to enquiries to chasing invoices — each with a clear line on what stays with you.
- Don’t try all ten. Pick the one problem that costs you most, test it for a month, and measure it.
Where things actually stand
In the US Federal Reserve’s survey of small employer firms, 46% said they used AI — but only 7% of those had it fully integrated into how the business runs. In the Royal Institution of Chartered Surveyors’ 2025 construction survey, about 45% of organisations had no AI in use at all, and fewer than 1% had it embedded across the business.
Neither of those is a measurement of Irish tradespeople. But they point the same way: the opportunity isn’t “AI transformation.” It’s finding one job that keeps going wrong, fixing it carefully, and checking it helped. (For the Irish picture, see how AI-ready Ireland’s trades businesses are.)
1. Get back to enquiries before they go elsewhere
You can’t always answer. A simple form or message flow can collect the basics — where, what job, how urgent, a few photos — so when you ring back you already know what it is. Customers should know it’s automated, and always have a way to reach you. It should never diagnose anything.
Read more: qualifying enquiries without losing the human touch and what missed calls could be costing you.
2. Chase invoices consistently
You can do great work and still be short at the end of the month because invoices go out late and reminders depend on you remembering. An assistant can list what’s coming due, draft a polite reminder and nudge you if nobody’s paid. It shouldn’t threaten, misstate your terms, or send anything you haven’t approved.
Read more: why a cash-flow assistant may matter more than a chatbot.
3. Win repeat work with aftercare reminders
A finished job often leads to the next one — an annual service, a seasonal check, a warranty reminder. The problem is remembering. A reminder system can tell you who’s due and draft a message from wording you’ve approved. Keep it relevant, easy to ignore, and within the rules on marketing messages.
Read more: could AI help small contractors win more repeat work?
4. Spot tenders and subcontracting work
Suitable work is scattered across public tender sites, buyer websites and trade contacts. A research assistant can watch the sources you choose and send a weekly shortlist filtered by trade, county and deadline. It doesn’t decide whether you’re qualified to bid — you check the original documents.
Read more: how AI could help small contractors find more work.
5. Turn voice notes into proper job records
You already talk into your phone in the van. The trouble is those notes stay buried. AI can turn one into a tidy record — customer, work done, materials, what’s outstanding, next step — for you to check. It should flag anything uncertain rather than fill in the blanks.
Read more: can voice notes replace the tradesperson’s notebook?
6. Sort photos, receipts and certificates
Useful records end up spread across your camera roll, WhatsApp, email and a folder in the van. AI can sort them by customer, address and job, and point out what’s missing — like a certificate that never got filed. These are your customers’ homes and details, so find out where any tool stores them first.
7. Plan the day with less driving
A scheduling assistant can group jobs by area, flag clashes and suggest a draft route. You still approve it, because it doesn’t know that one customer can only do mornings or that a street has nowhere to park. For a sole trader, simple reminders and rescheduling may be all you need.
Read more: could AI reduce wasted travel?
8. Keep the van stocked
One missing fitting can mean another trip to the merchants. An assistant can record what you used, suggest a restock list and spot what you keep buying in a hurry. It recommends; you order.
Read more: can AI make van stock easier to manage?
9. Draft site paperwork — and check it
Method statements, site reports and handover packs eat evenings. AI can fill in a first draft from your own templates and notes. The key word is draft: it must never invent safety controls or site conditions, and a competent person signs it off. Construction professionals surveyed by RICS rated reviewing contracts and documents (30%) and risk management (29%) among the areas where AI could make a big difference.
Read more: can AI reduce the burden of construction paperwork?
10. Support apprentices and new staff
AI can turn your own approved materials into revision questions, checklists and plain-English explanations. It supports supervision; it doesn’t replace it. No trainee should rely on a chatbot for a decision involving safety or an unfamiliar site.
Read more: could AI support apprentices without replacing experienced tradespeople?
The same rules apply to all ten
| Rule | In practice |
|---|---|
| Start narrow | One recurring job, not the whole business |
| You stay accountable | You approve prices, technical advice, safety information and promises |
| Show the gaps | It flags what’s missing or unclear instead of guessing |
| Protect people’s details | Know where information is stored, who sees it and how long it’s kept |
| Measure the result | Enquiries recovered, jobs booked, invoices paid, repeat work — not messages sent |
| Keep a record of mistakes | Note what went wrong and fix the process |
For more on why those limits matter, read the human side of AI in the trades. If quoting is your bottleneck, AI-assisted quoting covers where it helps and where it doesn’t.
Where to start
| If this sounds like you… | Start with |
|---|---|
| You miss calls while you’re working | Getting back to enquiries (1) |
| Plenty of work, slow to get paid | Chasing invoices (2) |
| Past customers never hear from you again | Aftercare reminders (3) |
| Your evenings go on notes and paperwork | Voice-note records (5) or paperwork drafts (9) |
| Several vans, days that keep falling apart | Scheduling (7) and van stock (8) |
Whatever you choose: note how things work now, pick one result to measure, run it for a month with your approval on everything that goes to a customer, then compare. Small enough to stop if it doesn’t help. And before paying anyone, read what AI is actually worth paying for.
What this research can’t tell you
The adoption figures above are from US small employers and from construction professionals worldwide, mostly at larger firms. No study I can find has tested any of these ten uses with Irish trades businesses, so none of them come with a promised saving or a guaranteed result. They’re ideas worth testing carefully, not proven products. You can see every source and its limits on how we check our sources.
Which one would help you most?
I’m running a short survey asking local tradespeople which of these everyday problems — calls, messages, quotes, follow-ups, invoices, reminders — they’d most like fixed. It takes 6–8 minutes, and leaving your contact details is optional.
Sources
- Federal Reserve Banks — 2026 Report on Employer Firms (Small Business Credit Survey, fielded September–November 2025)
- RICS — Artificial intelligence in construction report (2025)